Thoracic malignancies account for more deaths than prostate, breast and colorectal cancer combined. The most frequently diagnosed lung cancer subtypes is lung adenocarcinoma (ADC), which can metastasize rapidly to multiple vital organs. Despite recent advances in the genetic classification of lung cancers, the molecular and biological determinants of this aggressive clinical course remain unknown. The mammalian lungs are complex organs that require the specification of various epithelial cell types for proper homeostasis. By employing innovative computational and experimental approaches, we uncovered a unique link between differentiation specific gene expression patterns of certain lung epithelial cells and human ADC recurrences. In particular, we discovered a metastasis suppressor pathway associated with the cell fate transcription factors GATA6 and HOPX. We propose that these transcription factors cooperate to restrain multiple biological steps in the metastatic cascade. Moreover, we hypothesize that GATA6 and HOPX control a lineage-specific metastasis program that is selective for epithelial cells of the distal airways and the lug ADC subtype. To study this pathway, we will perform experiments that take advantage of spatio-temporally controlled gene gain or loss of function approaches in vivo. First, we will characterize the biological function(s) of GATA6 and HOPX during tumorigenesis, distant organ metastasis, and lung ADC differentiation using our established in vivo model of metastatic dissemination and colonization by human ADC cells. Second, we will develop a complementary model in genetically engineered mice, to study how loss of Gata6 and/or Hopx cooperates with Kras and p53 mutations at different stages of metastatic progression, and their associated effects on endogenous stem/progenitor cells of the murine airways. Finally, we will integrate multiple approaches to elucidate the downstream mechanism by which the GATA6/HOPX pathway restrains metastatic progression in human ADCs, through the control of their target gene, the Hsa-miR-302/367 cluster. Our project is predicted to reveal conserved epigenetic networks that control lung homeostasis, and how their perturbation endows thoracic cancers with metastatic competence to multiple tissues. The overall goal of our proposal may therefore shed new light into the origins of aggressive metastatic tumors, and provide insights into more accurate diagnostic modalities for lung cancer patients at risk for metastatic disease.